Skip to content

The Patron Saint of Superheroes

Chris Gavaler Explores the Multiverse of Comics, Pop Culture, and Politics

The one benefit of Virginia’s and Pennsylvania’s CV-19 lockdowns: daily study halls with my son at our dining room table. He’s a first-year at Haverford and would normally be enjoying his classes up in Philadelphia, and I’m on sabbatical and would normally be wallowing in the too-quiet empty nest of our house. I feel a kind of Monkey’s Paw guilt about the pandemic, since I had been not-so-secretly wishing my son were still living at home when the world as we knew it came to an unexpected end. Now Cameron and I can procrastinate on our laptops at opposite corners of the same table.

That includes my sliding this image across to him and asking:

“What order would you view these panels?”

It’s a page from Matt Baker’s “Sky Girl” in Jumbo Comics No. 83 (December 1945). I’m writing on Baker for Qiana Whitted’s collection Desegregating Comics: ​Debating Blackness in Early American Comics, 1900-1960. My chapter isn’t due till August, but like I said, sabbatical. Also, Matt Baker’s layouts are pleasantly insane. The content is too (note the so-called “Good Girl” art, including the impossibly small feet, impossibly thin waist, and impossibly long legs of the panel five figure), but my analysis is (mostly) of his disruptive viewing paths.

Which is why I wanted to see how Cameron navigated the page. According to Neil Cohn, who has done some of the most thorough analysis of viewing paths, viewers follow eight protocols:

    1. “Go to the left corner.”
    2. “If no top left panel, go to either the highest and/or leftmost panel.”
    3. “Follow the outer border.”
    4. “Follow the inner border.”
    5. “Move to the right.”
    6. “Move straight down.”
    7. “If nothing is to the right, go to the far left and down.”
    8. “Go to the panel that has not been read yet.”

But those don’t really explain Baker’s page. Numbering the panels according to the order determined by their content gets you this:

Comics theory doesn’t have much in the way of terminology for describing viewing paths. This page, like essentially all of Baker’s pages, follows an underlying Z-path and so the assumption that the images are arranged in rows rather than an N-path’s columns.

But that doesn’t explain the leap from panel 3 to panel 4–especially the unusual way it requires viewers to skip the not-yet-viewed panel 5. I’m calling that a “parallel saccade.” A saccade is the quick eye movement that usually happens at the end of a row and the start of a next row. A Z-path saccade usually moves diagonally left and down, but Baker’s instead moves in a parallel line.

Getting from panel 5 to 6 is easy enough, but it requires moving from right to left–the exact opposite of Z-path norms. I’m calling that a “reversed path,” because, well, that’s the most obvious name I could come up with. (If you have to coin a new term, you should make it as usefully self-evident as possible.)

Then moving from panel 6 to 7 requires another leap, this time over part of an image that has already been viewed. I’m calling that a “segment leap,” because the leap occurs not as part of a saccade (where it’s normal) but within the progression of a row, which is a kind of segment (allowing at least the theoretical possibility that it could happen within a column segment too).

Baker uses these three (and a couple more) disruptive techniques routinely. That’s why writing his chapter has been so much fun.

So when I slid my computer screen over to Cameron’s corner of the table, I thought he was going to get gummed up and have to pay careful attention to the image content, stopping and backing up a couple times to navigate the right order. But he didn’t. He got it on his first try, and it only took a few seconds, pausing only slightly at those three trouble spots. Which prompted my next question:

“How the hell did you do that?”

I was lucky that I got to watch him, because it was clear that he was doing something holistic and intuitive. Baker’s atypical techniques were virtually no challenge. Which suggests that Cohn’s protocols are missing something. Rather than assessing order according to the arrangement of entire panels by following frame and gutter edges, viewers seem to be doing something both simpler and more effective.

Cameron just attended to some small portion of each panel and ignored the rest. When placing numbers in my diagrams, I tried to approximate the geographic center of each, assuming centers are important. But apparently they’re not.

Look what happens when I instead delete everything but the top left corner of each panel:

All those weird Baker techniques vanish.

But look at what happens when I delete everything but the bottom right corners:

The confusion intensifies. 1, 2, 4, 3, 6, 5, 7 makes no sense. Which suggests that viewers don’t care about the supposed exit area of panels, only their top left entrance points. This rejects comics theorist Theirry Groensteen’s claim that a viewer’s eye “always arrives … from another point situated within the” viewing path and an “exit is always indicated, pointing to another” panel. It does’t matter where the image ends. A next panel is determined by the entrance area of the current panel.

I tested the approach on a few more Baker examples. These three pages combine some of his techniques for some particularly odd moves:

But entrance-order clarifies them:

All but the last two panels in the second diagrams, because it’s not clear if highest or leftmost gets ordering priority.

So this approach will still need some testing and de-bugging, but I think it does a better job of explaining what comics viewers actually do when navigating a page. When I showed this all to Cameron while we were still sitting at the table (a new chapter of his favorite blog novel had dropped that morning, so progress on his math homework was appropriately slow), he asked:

“So are you going to give me credit for this?”

Yes. I hereby name the system for navigating panel order according to image entrance areas:

Cameron Corners.

%d bloggers like this: